Analytics Market Chart

Measuring the Influence of Your Marketing


If you're relying on "conversions" to measure your marketing you're not seeing the whole picture.

One of the primary reasons for using Google Analytics is to make decisions about your marketing efforts. After analyzing reports, you should be able to answer the question, "What do I do now?" Do you stop a marketing campaign because it's not working? Do you beef up another one that is surprisingly successful?

To do this you must take some measurement of the relative influence of a campaign. Certainly, one useful measurement for this is conversion, but it's one-dimensional.

How Google Analytics Attributes Conversions

Every time a visitor comes to your site, the tracking code will try to determine where the visitor came from and store this information in their cookies. If the visitor comes back later via another source, it will overwrite the cookies with the new traffic source.

When a visitor converts (either completes a pre-defined goal or makes a purchase) Google Analytics will attribute the conversion to the source in the cookies.

In other words, if I come to your site through a banner ad and come back later through a Google search before converting, that conversion will be attributed only to the Google search.

An Incomplete Picture

We can argue all day about whether GA should do this differently and how. I submit, however, that relying too heavily on conversion metrics is a mistake. A silly, naive mistake, in fact.

Actually, I'll go further than that. Relying too heavily on any single metric is absurd.

We're talking about making decisions based on aggregate data of individual visitors making independent decisions during multiple visits and assuming that we know to what extent they were motivated by this or that. Think about it for a minute. It's a ridiculous presumption.

Instead, to make any sense of this, we have to look at it from multiple facets and try to get a complete picture from all the different variables.

Using Other Metrics

First, we should establish that when a visitor comes back to a site via some new source, their previous visit information is still stored in the reports. And it's still attributed to the previous traffic sources.

So, think about all the metrics that give us some indication that a marketing campaign was influential even if it didn't lead to an immediate sale. For example, these metrics spring to mind:

  • Time on site
  • Bounce rate
  • Time on landing page
  • Pages per visit

Other metrics may make sense for your specific website, too.

Now, look at all of your marketing and compare these numbers. Is there a specific campaign that has low conversion rates, low time on site, low pages per visit and high bounce rates? You can feel pretty confident that it is not very influential.

On the other hand a campaign with no conversions but with a high time on site, high pages per visit, high time on landing page and a low bounce rate was probably very influential in the conversion process. These metrics would represent an important--perhaps vital--early step toward conversion. Focusing too much on conversion numbers would never tell you that.

This is a more comprehensive approach to measuring the influence of your marketing.

An analyst who tries to tout the virtue of one all-powerful metric (conversion rate, clicks, ROI, etc.) is probably not a very good analyst. Or they're trying to sell you something.

However, if they're trying to sell you Girl Scout cookies, will you please send them my way?